An important aspect of developing treatments against cancer consists in correcting the defects caused by the abnormal activity of oncogenes and tumor suppressor genes. This usually includes the assignment of cancer- related gene products to their respective biochemical pathways. It has been shown that the genetics available in model organisms such as yeast, nematodes, flies, and mice can be highly advantageous for these projects. Despite the power of model organisms, the number of genes with a function assigned is relatively small and thus the functional analysis of cancer-related orthologs is still relatively tedious. However, with the complete sequence of the genome of most model organisms anticipated to be available soon, several laboratories have initiated the development of genome-wide gene-function analysis projects. Such projects, collectively referred to as "functional genomics" include genome-wide expression analysis, gene knock-outs and protein-protein interaction mapping. They are expected to reveal gene functions at a drastically increased rate. In this context, the long-term goal of our laboratory is to generate a comprehensive protein-protein interaction map for the nematode C. elegans. This will be achieved in three steps: Step I: validation of our improved version of the yeast two-hybrid system to generate protein interaction maps (funded by an R01 grant from the NHRGI); Step II: development of new high-throughput technologies for protein interaction mapping (object of this grant); step III: production phase (planned for the long-term and not the object of this grant). In step II, we plan to automate most steps of the two-hybrid methodology using a new cloning method, referred to as Gateway cloning. Gateway allows the transfer of DNA inserts between donor plasmids (or PCR products) and recipient plasmids. It is based on an in vitro site- specific recombination event mediated by purified phage lambda proteins and thus eliminates the need for restriction enzymes and ligases. It is fast and reversible, and the whole procedure can take place in 96- well plates, which means that automation is possible. In summary, Gateway will allow us to go in and come out of the two-hybrid in a completely automated series of steps. We will first develop the Gateway technology in the context of the two- hybrid system (R21). Then we will develop high-throughput methods for protein interaction mapping based on the use of the Gateway/two-hybrid system (R33). We will also adapt the technology so that other worm functional projects can benefit from it.